LLM as a Server – Overview
This site demonstrates a complete web‑application built solely from a language model. The runtime provides only deterministic transport, per‑session state storage, request parsing and a minimal virtual filesystem. All application logic – routing, content generation, persistence – lives in the model itself.
Contents
- Architecture
- State Model
- API Conventions
- Virtual Filesystem
- Limitations & Trust Boundary
- Demo: Create a Note
Each page focuses on a single topic and links back to the index, allowing a visitor to explore the system without ever loading a large monolithic document.